I believe when you end the process instance it's node isn't actually changed by design. It will just show you this "ended" state, but it will stay at it's current node. This is specially helpful in a situation where you have more than just one end state ("to which one should the token go?"). In case you have a complex process design between the start and end state (or at least a series of Task nodes and others) you can send the token to the end state directly changing it's node using the API (Token.setNode). This way you avoid creating all the task instances and stuff that could appear in the way, you just directly point the root token to the desired end node. Cristian www.miro.cl View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4054121#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...